This study will examine the internal and external forces that contribute to rural hospital affiliation with a multihospital system (MHS). The University of Michigan-Johns Hopkins Rural Hospital Survey (UMJHRHS) will supply hospital specific information such as the types and number of hospital services offered, strategic management activities undertaken (including MHS affiliation), and financial status, for a random sample of 797 U.S. rural (nonmetropolitan) hospitals for the years 1983-1988. In addition, county-specific environmental data obtained from the 1991 Area Resource Files will be matched with each hospital from the UMJHRHS to account for the various environmental pressures encountered by sample hospitals. Together, these data provide a rich source of information on rural hospitals and their local environments. The primary dependent variable for this study is whether or not sample hospitals joined a system between 1983-1988. For hypotheses related to specific MHS arrangements, the dependent variable will reflect whether system affiliation is through an owned/leased/sponsored or contract-managed arrangement. Appropriate bivariate and multivariate techniques will be used for data exploration. In addition, Cox's proportional hazards model will be used to account for right censoring, time-varying explanatory variables, and a dichotomous dependent variable. Results from this study will provide comprehensive information regarding the internal activities and environmental pressures faced by U.S. rural hospitals, and the resultant effects that those elements have on MHS affiliation. In addition, this study will supply important baseline information for future research investigating whether MHS affiliation improves rural hospital viability.